@alisio/plugin-semantic 0.0.0-stage → 0.1.0

package/dist/notebook/db.js

@@ -0,0 +1,328 @@
+/** Numbered migrations (§7.1); `PRAGMA user_version` records the last one applied. */
+const MIGRATIONS = [
+    `CREATE TABLE notebooks (
+     id TEXT PRIMARY KEY,
+     title TEXT NOT NULL,
+     created_at INTEGER NOT NULL,
+     updated_at INTEGER NOT NULL,
+     remote_notice_ack INTEGER NOT NULL DEFAULT 0
+   );
+   CREATE TABLE sources (
+     id TEXT PRIMARY KEY,
+     notebook_id TEXT NOT NULL,
+     name TEXT NOT NULL,
+     kind TEXT NOT NULL,
+     mime TEXT NOT NULL,
+     sha256 TEXT NOT NULL,
+     bytes INTEGER NOT NULL,
+     status TEXT NOT NULL,
+     error TEXT,
+     chunk_count INTEGER NOT NULL DEFAULT 0,
+     created_at INTEGER NOT NULL
+   );
+   CREATE INDEX sources_notebook ON sources(notebook_id);
+   CREATE TABLE chunks (
+     id INTEGER PRIMARY KEY,
+     source_id TEXT NOT NULL,
+     ordinal INTEGER NOT NULL,
+     text TEXT NOT NULL,
+     page INTEGER,
+     char_start INTEGER NOT NULL,
+     char_end INTEGER NOT NULL,
+     UNIQUE (source_id, ordinal)
+   );
+   CREATE VIRTUAL TABLE chunks_fts USING fts5(
+     text, content='chunks', content_rowid='id', tokenize='unicode61 remove_diacritics 2'
+   );
+   CREATE TRIGGER chunks_ai AFTER INSERT ON chunks BEGIN
+     INSERT INTO chunks_fts(rowid, text) VALUES (new.id, new.text);
+   END;
+   CREATE TRIGGER chunks_ad AFTER DELETE ON chunks BEGIN
+     INSERT INTO chunks_fts(chunks_fts, rowid, text) VALUES ('delete', old.id, old.text);
+   END;
+   CREATE TABLE messages (
+     id INTEGER PRIMARY KEY AUTOINCREMENT,
+     notebook_id TEXT NOT NULL,
+     role TEXT NOT NULL,
+     text TEXT NOT NULL,
+     citations_json TEXT,
+     created_at INTEGER NOT NULL
+   );
+   CREATE INDEX messages_notebook ON messages(notebook_id, id);
+   CREATE TABLE jobs (
+     id TEXT PRIMARY KEY,
+     notebook_id TEXT NOT NULL,
+     kind TEXT NOT NULL,
+     state TEXT NOT NULL,
+     progress REAL NOT NULL DEFAULT 0,
+     error TEXT,
+     result_json TEXT,
+     source_id TEXT,
+     created_at INTEGER NOT NULL,
+     updated_at INTEGER NOT NULL
+   );`,
+];
+const notebook = (r) => ({
+    id: r.id,
+    title: r.title,
+    createdAt: Number(r.created_at),
+    updatedAt: Number(r.updated_at),
+    remoteNoticeAck: Number(r.remote_notice_ack) === 1,
+});
+const source = (r) => ({
+    id: r.id,
+    notebookId: r.notebook_id,
+    name: r.name,
+    kind: r.kind,
+    mime: r.mime,
+    sha256: r.sha256,
+    bytes: Number(r.bytes),
+    status: r.status,
+    error: r.error ?? null,
+    chunkCount: Number(r.chunk_count),
+    createdAt: Number(r.created_at),
+});
+const chunk = (r) => ({
+    id: Number(r.id),
+    sourceId: r.source_id,
+    ordinal: Number(r.ordinal),
+    text: r.text,
+    page: r.page === null || r.page === undefined ? null : Number(r.page),
+    charStart: Number(r.char_start),
+    charEnd: Number(r.char_end),
+});
+const parse = (text) => {
+    if (typeof text !== "string")
+        return undefined;
+    try {
+        return JSON.parse(text);
+    }
+    catch {
+        return undefined;
+    }
+};
+const job = (r) => ({
+    id: r.id,
+    notebookId: r.notebook_id,
+    kind: r.kind,
+    state: r.state,
+    progress: Number(r.progress),
+    error: r.error ?? null,
+    result: parse(r.result_json) ?? null,
+    sourceId: r.source_id ?? null,
+    createdAt: Number(r.created_at),
+    updatedAt: Number(r.updated_at),
+});
+/** The Notebook's own SQLite store (D10): sources, chunks + FTS5, chat and jobs. */
+export class NotebookDb {
+    sql;
+    now;
+    constructor(sql, now = Date.now) {
+        this.sql = sql;
+        this.now = now;
+        this.migrate();
+        // Jobs that were running when the process stopped cannot resume.
+        this.sql
+            .prepare("UPDATE jobs SET state = 'failed', error = 'operation_failed', updated_at = ? WHERE state IN ('queued', 'running')")
+            .run(this.now());
+        this.sql
+            .prepare("UPDATE sources SET status = 'failed', error = 'operation_failed' WHERE status = 'indexing'")
+            .run();
+    }
+    migrate() {
+        const version = Number(this.sql.prepare("PRAGMA user_version").get()?.user_version ?? 0);
+        for (let v = version; v < MIGRATIONS.length; v++)
+            this.sql.transaction(() => {
+                this.sql.exec(MIGRATIONS[v]);
+                this.sql.exec(`PRAGMA user_version = ${v + 1}`);
+            });
+    }
+    close() {
+        this.sql.close();
+    }
+    // Notebooks
+    createNotebook(id, title) {
+        const at = this.now();
+        this.sql
+            .prepare("INSERT INTO notebooks (id, title, created_at, updated_at) VALUES (?, ?, ?, ?)")
+            .run(id, title, at, at);
+    }
+    notebook(id) {
+        const row = this.sql.prepare("SELECT * FROM notebooks WHERE id = ?").get(id);
+        return row ? notebook(row) : undefined;
+    }
+    notebooks() {
+        return this.sql
+            .prepare("SELECT n.*, (SELECT COUNT(*) FROM sources s WHERE s.notebook_id = n.id) AS source_count FROM notebooks n ORDER BY n.updated_at DESC, n.id")
+            .all()
+            .map((r) => ({ ...notebook(r), sourceCount: Number(r.source_count) }));
+    }
+    findNotebook(reference) {
+        const byId = this.notebook(reference);
+        if (byId)
+            return byId;
+        const row = this.sql
+            .prepare("SELECT * FROM notebooks WHERE title = ? COLLATE NOCASE ORDER BY updated_at DESC LIMIT 1")
+            .get(reference);
+        return row ? notebook(row) : undefined;
+    }
+    renameNotebook(id, title) {
+        this.sql
+            .prepare("UPDATE notebooks SET title = ?, updated_at = ? WHERE id = ?")
+            .run(title, this.now(), id);
+    }
+    touchNotebook(id) {
+        this.sql.prepare("UPDATE notebooks SET updated_at = ? WHERE id = ?").run(this.now(), id);
+    }
+    ackNotice(id) {
+        this.sql.prepare("UPDATE notebooks SET remote_notice_ack = 1 WHERE id = ?").run(id);
+    }
+    deleteNotebook(id) {
+        this.sql.transaction(() => {
+            this.sql
+                .prepare("DELETE FROM chunks WHERE source_id IN (SELECT id FROM sources WHERE notebook_id = ?)")
+                .run(id);
+            this.sql.prepare("DELETE FROM sources WHERE notebook_id = ?").run(id);
+            this.sql.prepare("DELETE FROM messages WHERE notebook_id = ?").run(id);
+            this.sql.prepare("DELETE FROM jobs WHERE notebook_id = ?").run(id);
+            this.sql.prepare("DELETE FROM notebooks WHERE id = ?").run(id);
+        });
+    }
+    countNotebooks() {
+        return Number(this.sql.prepare("SELECT COUNT(*) AS n FROM notebooks").get()?.n ?? 0);
+    }
+    // Sources
+    insertSource(row) {
+        this.sql
+            .prepare("INSERT INTO sources (id, notebook_id, name, kind, mime, sha256, bytes, status, created_at) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)")
+            .run(row.id, row.notebookId, row.name, row.kind, row.mime, row.sha256, row.bytes, row.status, this.now());
+    }
+    source(id) {
+        const row = this.sql.prepare("SELECT * FROM sources WHERE id = ?").get(id);
+        return row ? source(row) : undefined;
+    }
+    sources(notebookId) {
+        return this.sql
+            .prepare("SELECT * FROM sources WHERE notebook_id = ? ORDER BY created_at, id")
+            .all(notebookId)
+            .map(source);
+    }
+    sourceBySha(notebookId, sha256) {
+        const row = this.sql
+            .prepare("SELECT * FROM sources WHERE notebook_id = ? AND sha256 = ?")
+            .get(notebookId, sha256);
+        return row ? source(row) : undefined;
+    }
+    setSourceStatus(id, status, error = null) {
+        this.sql
+            .prepare("UPDATE sources SET status = ?, error = ? WHERE id = ?")
+            .run(status, error, id);
+    }
+    deleteSource(id) {
+        this.sql.transaction(() => {
+            this.sql.prepare("DELETE FROM chunks WHERE source_id = ?").run(id);
+            this.sql.prepare("DELETE FROM sources WHERE id = ?").run(id);
+        });
+    }
+    // Chunks (one transaction per source, D10)
+    replaceChunks(sourceId, chunks) {
+        this.sql.transaction(() => {
+            this.sql.prepare("DELETE FROM chunks WHERE source_id = ?").run(sourceId);
+            const insert = this.sql.prepare("INSERT INTO chunks (source_id, ordinal, text, page, char_start, char_end) VALUES (?, ?, ?, ?, ?, ?)");
+            for (const c of chunks)
+                insert.run(sourceId, c.ordinal, c.text, c.page, c.charStart, c.charEnd);
+            this.sql
+                .prepare("UPDATE sources SET chunk_count = ? WHERE id = ?")
+                .run(chunks.length, sourceId);
+        });
+    }
+    clearChunks(sourceId) {
+        this.replaceChunks(sourceId, []);
+    }
+    chunks(sourceId) {
+        return this.sql
+            .prepare("SELECT * FROM chunks WHERE source_id = ? ORDER BY ordinal")
+            .all(sourceId)
+            .map(chunk);
+    }
+    chunk(sourceId, ordinal) {
+        const row = this.sql
+            .prepare("SELECT * FROM chunks WHERE source_id = ? AND ordinal = ?")
+            .get(sourceId, ordinal);
+        return row ? chunk(row) : undefined;
+    }
+    notebookChunkCount(notebookId) {
+        return Number(this.sql
+            .prepare("SELECT COUNT(*) AS n FROM chunks c JOIN sources s ON s.id = c.source_id WHERE s.notebook_id = ?")
+            .get(notebookId)?.n ?? 0);
+    }
+    /** BM25 ranking over the notebook's chunks, as `<sourceId>:<ordinal>` keys. */
+    keywordSearch(notebookId, match, limit, sourceIds) {
+        if (!match)
+            return [];
+        const scope = sourceIds?.length
+            ? ` AND c.source_id IN (${sourceIds.map(() => "?").join(", ")})`
+            : "";
+        try {
+            return this.sql
+                .prepare(`SELECT c.source_id, c.ordinal FROM chunks_fts f JOIN chunks c ON c.id = f.rowid
+           JOIN sources s ON s.id = c.source_id
+           WHERE chunks_fts MATCH ? AND s.notebook_id = ?${scope}
+           ORDER BY bm25(chunks_fts), c.id LIMIT ?`)
+                .all(match, notebookId, ...(sourceIds ?? []), limit)
+                .map((r) => `${r.source_id}:${Number(r.ordinal)}`);
+        }
+        catch {
+            return [];
+        }
+    }
+    // Messages
+    addMessage(notebookId, role, text, meta = {}) {
+        const result = this.sql
+            .prepare("INSERT INTO messages (notebook_id, role, text, citations_json, created_at) VALUES (?, ?, ?, ?, ?)")
+            .run(notebookId, role, text, JSON.stringify(meta), this.now());
+        return Number(result.lastInsertRowid);
+    }
+    messages(notebookId, limit, before) {
+        const rows = this.sql
+            .prepare(`SELECT * FROM messages WHERE notebook_id = ?${before ? " AND id < ?" : ""} ORDER BY id DESC LIMIT ?`)
+            .all(...(before ? [notebookId, before, limit] : [notebookId, limit]));
+        return rows.reverse().map((r) => ({
+            id: Number(r.id),
+            notebookId: r.notebook_id,
+            role: r.role,
+            text: r.text,
+            meta: parse(r.citations_json) ?? {},
+            createdAt: Number(r.created_at),
+        }));
+    }
+    clearMessages(notebookId) {
+        this.sql.prepare("DELETE FROM messages WHERE notebook_id = ?").run(notebookId);
+    }
+    // Jobs
+    insertJob(id, notebookId, kind, sourceId) {
+        const at = this.now();
+        this.sql
+            .prepare("INSERT INTO jobs (id, notebook_id, kind, state, source_id, created_at, updated_at) VALUES (?, ?, ?, 'queued', ?, ?, ?)")
+            .run(id, notebookId, kind, sourceId, at, at);
+    }
+    updateJob(id, patch) {
+        const current = this.job(id);
+        if (!current)
+            return;
+        this.sql
+            .prepare("UPDATE jobs SET state = ?, progress = ?, error = ?, result_json = ?, updated_at = ? WHERE id = ?")
+            .run(patch.state ?? current.state, patch.progress ?? current.progress, patch.error === undefined ? current.error : patch.error, JSON.stringify(patch.result === undefined ? current.result : patch.result), this.now(), id);
+    }
+    job(id) {
+        const row = this.sql.prepare("SELECT * FROM jobs WHERE id = ?").get(id);
+        return row ? job(row) : undefined;
+    }
+    activeJobs(notebookId) {
+        return this.sql
+            .prepare("SELECT * FROM jobs WHERE notebook_id = ? AND state IN ('queued', 'running') ORDER BY created_at")
+            .all(notebookId)
+            .map(job);
+    }
+}
+//# sourceMappingURL=db.js.map

package/dist/notebook/db.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"db.js","sourceRoot":"","sources":["../../src/notebook/db.ts"],"names":[],"mappings":"AAGA,sFAAsF;AACtF,MAAM,UAAU,GAAa;IAC3B;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;MA4DI;CACL,CAAC;AAsDF,MAAM,QAAQ,GAAG,CAAC,CAAS,EAAe,EAAE,CAAC,CAAC;IAC5C,EAAE,EAAE,CAAC,CAAC,EAAE;IACR,KAAK,EAAE,CAAC,CAAC,KAAK;IACd,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;IAC/B,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;IAC/B,eAAe,EAAE,MAAM,CAAC,CAAC,CAAC,iBAAiB,CAAC,KAAK,CAAC;CACnD,CAAC,CAAC;AACH,MAAM,MAAM,GAAG,CAAC,CAAS,EAAa,EAAE,CAAC,CAAC;IACxC,EAAE,EAAE,CAAC,CAAC,EAAE;IACR,UAAU,EAAE,CAAC,CAAC,WAAW;IACzB,IAAI,EAAE,CAAC,CAAC,IAAI;IACZ,IAAI,EAAE,CAAC,CAAC,IAAI;IACZ,IAAI,EAAE,CAAC,CAAC,IAAI;IACZ,MAAM,EAAE,CAAC,CAAC,MAAM;IAChB,KAAK,EAAE,MAAM,CAAC,CAAC,CAAC,KAAK,CAAC;IACtB,MAAM,EAAE,CAAC,CAAC,MAAM;IAChB,KAAK,EAAE,CAAC,CAAC,KAAK,IAAI,IAAI;IACtB,UAAU,EAAE,MAAM,CAAC,CAAC,CAAC,WAAW,CAAC;IACjC,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;CAChC,CAAC,CAAC;AACH,MAAM,KAAK,GAAG,CAAC,CAAS,EAAY,EAAE,CAAC,CAAC;IACtC,EAAE,EAAE,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC;IAChB,QAAQ,EAAE,CAAC,CAAC,SAAS;IACrB,OAAO,EAAE,MAAM,CAAC,CAAC,CAAC,OAAO,CAAC;IAC1B,IAAI,EAAE,CAAC,CAAC,IAAI;IACZ,IAAI,EAAE,CAAC,CAAC,IAAI,KAAK,IAAI,IAAI,CAAC,CAAC,IAAI,KAAK,SAAS,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC;IACrE,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;IAC/B,OAAO,EAAE,MAAM,CAAC,CAAC,CAAC,QAAQ,CAAC;CAC5B,CAAC,CAAC;AACH,MAAM,KAAK,GAAG,CAAC,IAAa,EAAE,EAAE;IAC9B,IAAI,OAAO,IAAI,KAAK,QAAQ;QAAE,OAAO,SAAS,CAAC;IAC/C,IAAI,CAAC;QACH,OAAO,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAC1B,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,SAAS,CAAC;IACnB,CAAC;AACH,CAAC,CAAC;AACF,MAAM,GAAG,GAAG,CAAC,CAAS,EAAU,EAAE,CAAC,CAAC;IAClC,EAAE,EAAE,CAAC,CAAC,EAAE;IACR,UAAU,EAAE,CAAC,CAAC,WAAW;IACzB,IAAI,EAAE,CAAC,CAAC,IAAI;IACZ,KAAK,EAAE,CAAC,CAAC,KAAK;IACd,QAAQ,EAAE,MAAM,CAAC,CAAC,CAAC,QAAQ,CAAC;IAC5B,KAAK,EAAE,CAAC,CAAC,KAAK,IAAI,IAAI;IACtB,MAAM,EAAE,KAAK,CAAC,CAAC,CAAC,WAAW,CAAC,IAAI,IAAI;IACpC,QAAQ,EAAE,CAAC,CAAC,SAAS,IAAI,IAAI;IAC7B,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;IAC/B,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;CAChC,CAAC,CAAC;AAEH,oFAAoF;AACpF,MAAM,OAAO,UAAU;IACZ,GAAG,CAAc;IACT,GAAG,CAAe;IAEnC,YAAY,GAAgB,EAAE,GAAG,GAAiB,IAAI,CAAC,GAAG;QACxD,IAAI,CAAC,GAAG,GAAG,GAAG,CAAC;QACf,IAAI,CAAC,GAAG,GAAG,GAAG,CAAC;QACf,IAAI,CAAC,OAAO,EAAE,CAAC;QACf,iEAAiE;QACjE,IAAI,CAAC,GAAG;aACL,OAAO,CACN,mHAAmH,CACpH;aACA,GAAG,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC,CAAC;QACnB,IAAI,CAAC,GAAG;aACL,OAAO,CACN,4FAA4F,CAC7F;aACA,GAAG,EAAE,CAAC;IACX,CAAC;IAEO,OAAO;QACb,MAAM,OAAO,GAAG,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,qBAAqB,CAAC,CAAC,GAAG,EAAE,EAAE,YAAY,IAAI,CAAC,CAAC,CAAC;QACzF,KAAK,IAAI,CAAC,GAAG,OAAO,EAAE,CAAC,GAAG,UAAU,CAAC,MAAM,EAAE,CAAC,EAAE;YAC9C,IAAI,CAAC,GAAG,CAAC,WAAW,CAAC,GAAG,EAAE;gBACxB,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC,CAAW,CAAC,CAAC;gBACvC,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;YAClD,CAAC,CAAC,CAAC;IACP,CAAC;IAED,KAAK;QACH,IAAI,CAAC,GAAG,CAAC,KAAK,EAAE,CAAC;IACnB,CAAC;IAED,YAAY;IACZ,cAAc,CAAC,EAAU,EAAE,KAAa;QACtC,MAAM,EAAE,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QACtB,IAAI,CAAC,GAAG;aACL,OAAO,CAAC,+EAA+E,CAAC;aACxF,GAAG,CAAC,EAAE,EAAE,KAAK,EAAE,EAAE,EAAE,EAAE,CAAC,CAAC;IAC5B,CAAC;IACD,QAAQ,CAAC,EAAU;QACjB,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,sCAAsC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAC7E,OAAO,GAAG,CAAC,CAAC,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;IACzC,CAAC;IACD,SAAS;QACP,OAAO,IAAI,CAAC,GAAG;aACZ,OAAO,CACN,2IAA2I,CAC5I;aACA,GAAG,EAAE;aACL,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,GAAG,QAAQ,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,MAAM,CAAC,CAAC,CAAC,YAAY,CAAC,EAAE,CAAC,CAAC,CAAC;IAC3E,CAAC;IACD,YAAY,CAAC,SAAiB;QAC5B,MAAM,IAAI,GAAG,IAAI,CAAC,QAAQ,CAAC,SAAS,CAAC,CAAC;QACtC,IAAI,IAAI;YAAE,OAAO,IAAI,CAAC;QACtB,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG;aACjB,OAAO,CACN,yFAAyF,CAC1F;aACA,GAAG,CAAC,SAAS,CAAC,CAAC;QAClB,OAAO,GAAG,CAAC,CAAC,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;IACzC,CAAC;IACD,cAAc,CAAC,EAAU,EAAE,KAAa;QACtC,IAAI,CAAC,GAAG;aACL,OAAO,CAAC,6DAA6D,CAAC;aACtE,GAAG,CAAC,KAAK,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,EAAE,CAAC,CAAC;IAChC,CAAC;IACD,aAAa,CAAC,EAAU;QACtB,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,kDAAkD,CAAC,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,EAAE,EAAE,EAAE,CAAC,CAAC;IAC3F,CAAC;IACD,SAAS,CAAC,EAAU;QAClB,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,yDAAyD,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;IACtF,CAAC;IACD,cAAc,CAAC,EAAU;QACvB,IAAI,CAAC,GAAG,CAAC,WAAW,CAAC,GAAG,EAAE;YACxB,IAAI,CAAC,GAAG;iBACL,OAAO,CACN,sFAAsF,CACvF;iBACA,GAAG,CAAC,EAAE,CAAC,CAAC;YACX,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,2CAA2C,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;YACtE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,4CAA4C,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;YACvE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,wCAAwC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;YACnE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,oCAAoC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QACjE,CAAC,CAAC,CAAC;IACL,CAAC;IACD,cAAc;QACZ,OAAO,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,qCAAqC,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,IAAI,CAAC,CAAC,CAAC;IACvF,CAAC;IAED,UAAU;IACV,YAAY,CAAC,GAA0D;QACrE,IAAI,CAAC,GAAG;aACL,OAAO,CACN,+HAA+H,CAChI;aACA,GAAG,CACF,GAAG,CAAC,EAAE,EACN,GAAG,CAAC,UAAU,EACd,GAAG,CAAC,IAAI,EACR,GAAG,CAAC,IAAI,EACR,GAAG,CAAC,IAAI,EACR,GAAG,CAAC,MAAM,EACV,GAAG,CAAC,KAAK,EACT,GAAG,CAAC,MAAM,EACV,IAAI,CAAC,GAAG,EAAE,CACX,CAAC;IACN,CAAC;IACD,MAAM,CAAC,EAAU;QACf,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,oCAAoC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAC3E,OAAO,GAAG,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;IACvC,CAAC;IACD,OAAO,CAAC,UAAkB;QACxB,OAAO,IAAI,CAAC,GAAG;aACZ,OAAO,CAAC,qEAAqE,CAAC;aAC9E,GAAG,CAAC,UAAU,CAAC;aACf,GAAG,CAAC,MAAM,CAAC,CAAC;IACjB,CAAC;IACD,WAAW,CAAC,UAAkB,EAAE,MAAc;QAC5C,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG;aACjB,OAAO,CAAC,4DAA4D,CAAC;aACrE,GAAG,CAAC,UAAU,EAAE,MAAM,CAAC,CAAC;QAC3B,OAAO,GAAG,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;IACvC,CAAC;IACD,eAAe,CAAC,EAAU,EAAE,MAAoB,EAAE,KAAK,GAAkB,IAAI;QAC3E,IAAI,CAAC,GAAG;aACL,OAAO,CAAC,uDAAuD,CAAC;aAChE,GAAG,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAAC,CAAC;IAC5B,CAAC;IACD,YAAY,CAAC,EAAU;QACrB,IAAI,CAAC,GAAG,CAAC,WAAW,CAAC,GAAG,EAAE;YACxB,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,wCAAwC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;YACnE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,kCAAkC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAC/D,CAAC,CAAC,CAAC;IACL,CAAC;IAED,2CAA2C;IAC3C,aAAa,CAAC,QAAgB,EAAE,MAAe;QAC7C,IAAI,CAAC,GAAG,CAAC,WAAW,CAAC,GAAG,EAAE;YACxB,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,wCAAwC,CAAC,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;YACzE,MAAM,MAAM,GAAG,IAAI,CAAC,GAAG,CAAC,OAAO,CAC7B,qGAAqG,CACtG,CAAC;YACF,KAAK,MAAM,CAAC,IAAI,MAAM;gBACpB,MAAM,CAAC,GAAG,CAAC,QAAQ,EAAE,CAAC,CAAC,OAAO,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,SAAS,EAAE,CAAC,CAAC,OAAO,CAAC,CAAC;YAC1E,IAAI,CAAC,GAAG;iBACL,OAAO,CAAC,iDAAiD,CAAC;iBAC1D,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,QAAQ,CAAC,CAAC;QAClC,CAAC,CAAC,CAAC;IACL,CAAC;IACD,WAAW,CAAC,QAAgB;QAC1B,IAAI,CAAC,aAAa,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC;IACnC,CAAC;IACD,MAAM,CAAC,QAAgB;QACrB,OAAO,IAAI,CAAC,GAAG;aACZ,OAAO,CAAC,2DAA2D,CAAC;aACpE,GAAG,CAAC,QAAQ,CAAC;aACb,GAAG,CAAC,KAAK,CAAC,CAAC;IAChB,CAAC;IACD,KAAK,CAAC,QAAgB,EAAE,OAAe;QACrC,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG;aACjB,OAAO,CAAC,0DAA0D,CAAC;aACnE,GAAG,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC;QAC1B,OAAO,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;IACtC,CAAC;IACD,kBAAkB,CAAC,UAAkB;QACnC,OAAO,MAAM,CACX,IAAI,CAAC,GAAG;aACL,OAAO,CACN,iGAAiG,CAClG;aACA,GAAG,CAAC,UAAU,CAAC,EAAE,CAAC,IAAI,CAAC,CAC3B,CAAC;IACJ,CAAC;IACD,+EAA+E;IAC/E,aAAa,CAAC,UAAkB,EAAE,KAAa,EAAE,KAAa,EAAE,SAAoB;QAClF,IAAI,CAAC,KAAK;YAAE,OAAO,EAAE,CAAC;QACtB,MAAM,KAAK,GAAG,SAAS,EAAE,MAAM;YAC7B,CAAC,CAAC,wBAAwB,SAAS,CAAC,GAAG,CAAC,GAAG,EAAE,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG;YAChE,CAAC,CAAC,EAAE,CAAC;QACP,IAAI,CAAC;YACH,OAAO,IAAI,CAAC,GAAG;iBACZ,OAAO,CACN;;2DAEiD,KAAK;mDACb,CAC1C;iBACA,GAAG,CAAC,KAAK,EAAE,UAAU,EAAE,GAAG,CAAC,SAAS,IAAI,EAAE,CAAC,EAAE,KAAK,CAAC;iBACnD,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,GAAG,CAAC,CAAC,SAAS,IAAI,MAAM,CAAC,CAAC,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC;QACvD,CAAC;QAAC,MAAM,CAAC;YACP,OAAO,EAAE,CAAC;QACZ,CAAC;IACH,CAAC;IAED,WAAW;IACX,UAAU,CACR,UAAkB,EAClB,IAA0B,EAC1B,IAAY,EACZ,IAAI,GAA4B,EAAE;QAElC,MAAM,MAAM,GAAG,IAAI,CAAC,GAAG;aACpB,OAAO,CACN,mGAAmG,CACpG;aACA,GAAG,CAAC,UAAU,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,EAAE,IAAI,CAAC,GAAG,EAAE,CAAC,CAAC;QACjE,OAAO,MAAM,CAAC,MAAM,CAAC,eAAe,CAAC,CAAC;IACxC,CAAC;IACD,QAAQ,CAAC,UAAkB,EAAE,KAAa,EAAE,MAAe;QACzD,MAAM,IAAI,GAAG,IAAI,CAAC,GAAG;aAClB,OAAO,CACN,+CAA+C,MAAM,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,EAAE,2BAA2B,CACtG;aACA,GAAG,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,UAAU,EAAE,MAAM,EAAE,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,UAAU,EAAE,KAAK,CAAC,CAAC,CAAC,CAAC;QACxE,OAAO,IAAI,CAAC,OAAO,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;YAChC,EAAE,EAAE,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC;YAChB,UAAU,EAAE,CAAC,CAAC,WAAW;YACzB,IAAI,EAAE,CAAC,CAAC,IAAI;YACZ,IAAI,EAAE,CAAC,CAAC,IAAI;YACZ,IAAI,EAAE,KAAK,CAAC,CAAC,CAAC,cAAc,CAAC,IAAI,EAAE;YACnC,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,UAAU,CAAC;SAChC,CAAC,CAAC,CAAC;IACN,CAAC;IACD,aAAa,CAAC,UAAkB;QAC9B,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,4CAA4C,CAAC,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;IACjF,CAAC;IAED,OAAO;IACP,SAAS,CAAC,EAAU,EAAE,UAAkB,EAAE,IAAY,EAAE,QAAuB;QAC7E,MAAM,EAAE,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QACtB,IAAI,CAAC,GAAG;aACL,OAAO,CACN,wHAAwH,CACzH;aACA,GAAG,CAAC,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,QAAQ,EAAE,EAAE,EAAE,EAAE,CAAC,CAAC;IACjD,CAAC;IACD,SAAS,CACP,EAAU,EACV,KAAuF;QAEvF,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAC7B,IAAI,CAAC,OAAO;YAAE,OAAO;QACrB,IAAI,CAAC,GAAG;aACL,OAAO,CACN,kGAAkG,CACnG;aACA,GAAG,CACF,KAAK,CAAC,KAAK,IAAI,OAAO,CAAC,KAAK,EAC5B,KAAK,CAAC,QAAQ,IAAI,OAAO,CAAC,QAAQ,EAClC,KAAK,CAAC,KAAK,KAAK,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,KAAK,EACvD,IAAI,CAAC,SAAS,CAAC,KAAK,CAAC,MAAM,KAAK,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,KAAK,CAAC,MAAM,CAAC,EAC1E,IAAI,CAAC,GAAG,EAAE,EACV,EAAE,CACH,CAAC;IACN,CAAC;IACD,GAAG,CAAC,EAAU;QACZ,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,iCAAiC,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QACxE,OAAO,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;IACpC,CAAC;IACD,UAAU,CAAC,UAAkB;QAC3B,OAAO,IAAI,CAAC,GAAG;aACZ,OAAO,CACN,iGAAiG,CAClG;aACA,GAAG,CAAC,UAAU,CAAC;aACf,GAAG,CAAC,GAAG,CAAC,CAAC;IACd,CAAC;CACF"}

package/dist/notebook/docx.d.ts

@@ -0,0 +1,3 @@
+export { decodeXml } from "./zip.js";
+export declare function docxText(bytes: Buffer): string;
+//# sourceMappingURL=docx.d.ts.map

package/dist/notebook/docx.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"docx.d.ts","sourceRoot":"","sources":["../../src/notebook/docx.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,SAAS,EAAE,MAAM,UAAU,CAAC;AAErC,wBAAgB,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAU9C"}

package/dist/notebook/docx.js

@@ -0,0 +1,16 @@
+// .docx text (§7.3): `word/document.xml`, `<w:p>` per paragraph, `<w:t>` runs, tabs and breaks.
+import { DOCX_LIMITS, decodeXml, ZipArchive } from "./zip.js";
+export { decodeXml } from "./zip.js";
+export function docxText(bytes) {
+    const xml = new ZipArchive(bytes, DOCX_LIMITS).xml("word/document.xml");
+    const paragraphs = [];
+    for (const match of xml.matchAll(/<w:p[\s>][\s\S]*?<\/w:p>|<w:p\/>/g)) {
+        let text = "";
+        for (const token of match[0].matchAll(/<w:t(?:\s[^>]*)?>([\s\S]*?)<\/w:t>|<w:tab\/>|<w:br\/>/g))
+            text += token[1] !== undefined ? decodeXml(token[1]) : token[0] === "<w:tab/>" ? "\t" : "\n";
+        if (text.trim())
+            paragraphs.push(text);
+    }
+    return paragraphs.join("\n\n");
+}
+//# sourceMappingURL=docx.js.map

package/dist/notebook/docx.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"docx.js","sourceRoot":"","sources":["../../src/notebook/docx.ts"],"names":[],"mappings":"AAAA,gGAAgG;AAChG,OAAO,EAAE,WAAW,EAAE,SAAS,EAAE,UAAU,EAAE,MAAM,UAAU,CAAC;AAE9D,OAAO,EAAE,SAAS,EAAE,MAAM,UAAU,CAAC;AAErC,MAAM,UAAU,QAAQ,CAAC,KAAa;IACpC,MAAM,GAAG,GAAG,IAAI,UAAU,CAAC,KAAK,EAAE,WAAW,CAAC,CAAC,GAAG,CAAC,mBAAmB,CAAC,CAAC;IACxE,MAAM,UAAU,GAAa,EAAE,CAAC;IAChC,KAAK,MAAM,KAAK,IAAI,GAAG,CAAC,QAAQ,CAAC,mCAAmC,CAAC,EAAE,CAAC;QACtE,IAAI,IAAI,GAAG,EAAE,CAAC;QACd,KAAK,MAAM,KAAK,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,wDAAwD,CAAC;YAC7F,IAAI,IAAI,KAAK,CAAC,CAAC,CAAC,KAAK,SAAS,CAAC,CAAC,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,KAAK,UAAU,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC;QAC/F,IAAI,IAAI,CAAC,IAAI,EAAE;YAAE,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACzC,CAAC;IACD,OAAO,UAAU,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;AACjC,CAAC"}

package/dist/notebook/extract.d.ts

@@ -0,0 +1,11 @@
+import type { Segment } from "./chunk.js";
+export type SourceFormat = "text" | "html" | "docx" | "pptx" | "xlsx" | "pdf";
+/** Format by file extension (case-insensitive), or undefined when unsupported. */
+export declare function formatOf(name: string): SourceFormat | undefined;
+export declare function htmlText(html: string): string;
+/** Text segments of an uploaded file (PDF text comes from the frame, see `extractPdfPages`). */
+export declare function extractBytes(name: string, bytes: Buffer): Promise<Segment[]>;
+export declare const PDF_MIN_CHARS_PER_PAGE = 20;
+/** One segment per PDF page; fewer than 20 characters per page on average is `no_text`. */
+export declare function extractPdfPages(pages: string[]): Segment[];
+//# sourceMappingURL=extract.d.ts.map

package/dist/notebook/extract.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"extract.d.ts","sourceRoot":"","sources":["../../src/notebook/extract.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,OAAO,EAAE,MAAM,YAAY,CAAC;AAI1C,MAAM,MAAM,YAAY,GAAG,MAAM,GAAG,MAAM,GAAG,MAAM,GAAG,MAAM,GAAG,MAAM,GAAG,KAAK,CAAC;AAgB9E,kFAAkF;AAClF,wBAAgB,QAAQ,CAAC,IAAI,EAAE,MAAM,GAAG,YAAY,GAAG,SAAS,CAG/D;AAcD,wBAAgB,QAAQ,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAU7C;AAiCD,gGAAgG;AAChG,wBAAsB,YAAY,CAAC,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,GAAG,OAAO,CAAC,OAAO,EAAE,CAAC,CAiBlF;AAED,eAAO,MAAM,sBAAsB,KAAK,CAAC;AAEzC,2FAA2F;AAC3F,wBAAgB,eAAe,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,EAAE,CAK1D"}

package/dist/notebook/extract.js

@@ -0,0 +1,99 @@
+import { SemanticPluginError } from "../errors.js";
+import { decodeXml, docxText } from "./docx.js";
+import { pptxSegments, xlsxSegments } from "./ooxml.js";
+const FORMATS = {
+    txt: "text",
+    md: "text",
+    markdown: "text",
+    csv: "text",
+    json: "text",
+    html: "html",
+    htm: "html",
+    docx: "docx",
+    pptx: "pptx",
+    xlsx: "xlsx",
+    pdf: "pdf",
+};
+/** Format by file extension (case-insensitive), or undefined when unsupported. */
+export function formatOf(name) {
+    const ext = /\.([A-Za-z0-9]+)$/.exec(name)?.[1]?.toLowerCase();
+    return ext ? FORMATS[ext] : undefined;
+}
+function utf8(bytes) {
+    try {
+        // The decoder strips a leading BOM by default.
+        return new TextDecoder("utf-8", { fatal: true }).decode(bytes);
+    }
+    catch {
+        throw new SemanticPluginError("invalid_encoding");
+    }
+}
+const BLOCK = /<\/?(?:p|div|br|h[1-6]|li|ul|ol|tr|table|section|article|header|footer|blockquote|pre)\b[^>]*>/gi;
+export function htmlText(html) {
+    const text = html
+        .replace(/<!--[\s\S]*?-->/g, "")
+        .replace(/<(script|style|noscript|template)\b[\s\S]*?<\/\1\s*>/gi, "")
+        .replace(BLOCK, "\n")
+        .replace(/<[^>]*>/g, "");
+    return decodeXml(decodeHtmlNamed(text))
+        .replace(/[ \t]+\n/g, "\n")
+        .replace(/\n{3,}/g, "\n\n")
+        .trim();
+}
+const NAMED = {
+    nbsp: " ",
+    eacute: "é",
+    aacute: "á",
+    iacute: "í",
+    oacute: "ó",
+    uacute: "ú",
+    ntilde: "ñ",
+    Eacute: "É",
+    Aacute: "Á",
+    Iacute: "Í",
+    Oacute: "Ó",
+    Uacute: "Ú",
+    Ntilde: "Ñ",
+    uuml: "ü",
+    copy: "©",
+    reg: "®",
+    hellip: "…",
+    mdash: "—",
+    ndash: "–",
+    laquo: "«",
+    raquo: "»",
+    iquest: "¿",
+    iexcl: "¡",
+    ccedil: "ç",
+    agrave: "à",
+    egrave: "è",
+};
+const decodeHtmlNamed = (text) => text.replace(/&([A-Za-z]+);/g, (match, name) => NAMED[name] ?? match);
+/** Text segments of an uploaded file (PDF text comes from the frame, see `extractPdfPages`). */
+export async function extractBytes(name, bytes) {
+    const format = formatOf(name);
+    switch (format) {
+        case "text":
+            return [{ text: utf8(bytes), page: null }];
+        case "html":
+            return [{ text: htmlText(utf8(bytes)), page: null }];
+        case "docx":
+            return [{ text: docxText(bytes), page: null }];
+        case "pptx":
+            return pptxSegments(bytes);
+        case "xlsx":
+            return xlsxSegments(bytes);
+        default:
+            // PDF text is extracted in the frame with PDF.js (§7.3); anything else is unsupported.
+            throw new SemanticPluginError("unsupported_format");
+    }
+}
+export const PDF_MIN_CHARS_PER_PAGE = 20;
+/** One segment per PDF page; fewer than 20 characters per page on average is `no_text`. */
+export function extractPdfPages(pages) {
+    const characters = pages.reduce((sum, page) => sum + page.trim().length, 0);
+    if (pages.length === 0 || characters / pages.length < PDF_MIN_CHARS_PER_PAGE)
+        throw new SemanticPluginError("no_text");
+    return pages.map((text, index) => ({ text, page: index + 1 }));
+}
+//# sourceMappingURL=extract.js.map

package/dist/notebook/extract.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"extract.js","sourceRoot":"","sources":["../../src/notebook/extract.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,mBAAmB,EAAE,MAAM,cAAc,CAAC;AAEnD,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,MAAM,WAAW,CAAC;AAChD,OAAO,EAAE,YAAY,EAAE,YAAY,EAAE,MAAM,YAAY,CAAC;AAIxD,MAAM,OAAO,GAAiC;IAC5C,GAAG,EAAE,MAAM;IACX,EAAE,EAAE,MAAM;IACV,QAAQ,EAAE,MAAM;IAChB,GAAG,EAAE,MAAM;IACX,IAAI,EAAE,MAAM;IACZ,IAAI,EAAE,MAAM;IACZ,GAAG,EAAE,MAAM;IACX,IAAI,EAAE,MAAM;IACZ,IAAI,EAAE,MAAM;IACZ,IAAI,EAAE,MAAM;IACZ,GAAG,EAAE,KAAK;CACX,CAAC;AAEF,kFAAkF;AAClF,MAAM,UAAU,QAAQ,CAAC,IAAY;IACnC,MAAM,GAAG,GAAG,mBAAmB,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,CAAC;IAC/D,OAAO,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;AACxC,CAAC;AAED,SAAS,IAAI,CAAC,KAAiB;IAC7B,IAAI,CAAC;QACH,+CAA+C;QAC/C,OAAO,IAAI,WAAW,CAAC,OAAO,EAAE,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC;IACjE,CAAC;IAAC,MAAM,CAAC;QACP,MAAM,IAAI,mBAAmB,CAAC,kBAAkB,CAAC,CAAC;IACpD,CAAC;AACH,CAAC;AAED,MAAM,KAAK,GACT,kGAAkG,CAAC;AAErG,MAAM,UAAU,QAAQ,CAAC,IAAY;IACnC,MAAM,IAAI,GAAG,IAAI;SACd,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC;SAC/B,OAAO,CAAC,wDAAwD,EAAE,EAAE,CAAC;SACrE,OAAO,CAAC,KAAK,EAAE,IAAI,CAAC;SACpB,OAAO,CAAC,UAAU,EAAE,EAAE,CAAC,CAAC;IAC3B,OAAO,SAAS,CAAC,eAAe,CAAC,IAAI,CAAC,CAAC;SACpC,OAAO,CAAC,WAAW,EAAE,IAAI,CAAC;SAC1B,OAAO,CAAC,SAAS,EAAE,MAAM,CAAC;SAC1B,IAAI,EAAE,CAAC;AACZ,CAAC;AAED,MAAM,KAAK,GAA2B;IACpC,IAAI,EAAE,GAAG;IACT,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,IAAI,EAAE,GAAG;IACT,IAAI,EAAE,GAAG;IACT,GAAG,EAAE,GAAG;IACR,MAAM,EAAE,GAAG;IACX,KAAK,EAAE,GAAG;IACV,KAAK,EAAE,GAAG;IACV,KAAK,EAAE,GAAG;IACV,KAAK,EAAE,GAAG;IACV,MAAM,EAAE,GAAG;IACX,KAAK,EAAE,GAAG;IACV,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;IACX,MAAM,EAAE,GAAG;CACZ,CAAC;AACF,MAAM,eAAe,GAAG,CAAC,IAAY,EAAE,EAAE,CACvC,IAAI,CAAC,OAAO,CAAC,gBAAgB,EAAE,CAAC,KAAK,EAAE,IAAY,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,IAAI,KAAK,CAAC,CAAC;AAEhF,gGAAgG;AAChG,MAAM,CAAC,KAAK,UAAU,YAAY,CAAC,IAAY,EAAE,KAAa;IAC5D,MAAM,MAAM,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC;IAC9B,QAAQ,MAAM,EAAE,CAAC;QACf,KAAK,MAAM;YACT,OAAO,CAAC,EAAE,IAAI,EAAE,IAAI,CAAC,KAAK,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,CAAC,CAAC;QAC7C,KAAK,MAAM;YACT,OAAO,CAAC,EAAE,IAAI,EAAE,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,CAAC,CAAC;QACvD,KAAK,MAAM;YACT,OAAO,CAAC,EAAE,IAAI,EAAE,QAAQ,CAAC,KAAK,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,CAAC,CAAC;QACjD,KAAK,MAAM;YACT,OAAO,YAAY,CAAC,KAAK,CAAC,CAAC;QAC7B,KAAK,MAAM;YACT,OAAO,YAAY,CAAC,KAAK,CAAC,CAAC;QAC7B;YACE,uFAAuF;YACvF,MAAM,IAAI,mBAAmB,CAAC,oBAAoB,CAAC,CAAC;IACxD,CAAC;AACH,CAAC;AAED,MAAM,CAAC,MAAM,sBAAsB,GAAG,EAAE,CAAC;AAEzC,2FAA2F;AAC3F,MAAM,UAAU,eAAe,CAAC,KAAe;IAC7C,MAAM,UAAU,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,IAAI,EAAE,EAAE,CAAC,GAAG,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC;IAC5E,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,IAAI,UAAU,GAAG,KAAK,CAAC,MAAM,GAAG,sBAAsB;QAC1E,MAAM,IAAI,mBAAmB,CAAC,SAAS,CAAC,CAAC;IAC3C,OAAO,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,KAAK,EAAE,EAAE,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,GAAG,CAAC,EAAE,CAAC,CAAC,CAAC;AACjE,CAAC"}

package/dist/notebook/jobs.d.ts

@@ -0,0 +1,31 @@
+import type { NotebookDb } from "./db.js";
+export interface JobContext {
+    signal: AbortSignal;
+    progress(value: number): void;
+}
+export type JobWork = (context: JobContext) => Promise<unknown>;
+export interface JobOptions {
+    /** Runs after a cancellation, before the job is marked `cancelled` (cleanup of partial work). */
+    onCancel?: () => Promise<void> | void;
+    /** Runs after a failure with its code (for example to mark a source as failed). */
+    onFail?: (code: string) => Promise<void> | void;
+}
+/**
+ * Long work as jobs (D11): the action returns a job id at once and the frame polls `jobs.get`.
+ * Jobs of one notebook run one after another; each has its own `AbortController`.
+ */
+export declare class JobRunner {
+    private readonly db;
+    private readonly controllers;
+    private readonly chains;
+    constructor(db: NotebookDb);
+    start(notebookId: string, kind: string, sourceId: string | null, work: JobWork, options?: JobOptions): `${string}-${string}-${string}-${string}-${string}`;
+    cancel(id: string): boolean;
+    busy(): boolean;
+    /** Cancels every job (plugin dispose). */
+    dispose(): void;
+    /** Resolves when every job queued so far has settled (tests and dispose). */
+    idle(): Promise<void>;
+    private execute;
+}
+//# sourceMappingURL=jobs.d.ts.map

package/dist/notebook/jobs.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"jobs.d.ts","sourceRoot":"","sources":["../../src/notebook/jobs.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,SAAS,CAAC;AAE1C,MAAM,WAAW,UAAU;IACzB,MAAM,EAAE,WAAW,CAAC;IACpB,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,IAAI,CAAC;CAC/B;AACD,MAAM,MAAM,OAAO,GAAG,CAAC,OAAO,EAAE,UAAU,KAAK,OAAO,CAAC,OAAO,CAAC,CAAC;AAChE,MAAM,WAAW,UAAU;IACzB,iGAAiG;IACjG,QAAQ,CAAC,EAAE,MAAM,OAAO,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC;IACtC,mFAAmF;IACnF,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,OAAO,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC;CACjD;AAED;;;GAGG;AACH,qBAAa,SAAS;IACpB,OAAO,CAAC,QAAQ,CAAC,EAAE,CAAa;IAChC,OAAO,CAAC,QAAQ,CAAC,WAAW,CAAsC;IAClE,OAAO,CAAC,QAAQ,CAAC,MAAM,CAAoC;IAE3D,YAAY,EAAE,EAAE,UAAU,EAEzB;IAED,KAAK,CACH,UAAU,EAAE,MAAM,EAClB,IAAI,EAAE,MAAM,EACZ,QAAQ,EAAE,MAAM,GAAG,IAAI,EACvB,IAAI,EAAE,OAAO,EACb,OAAO,GAAE,UAAe,uDAazB;IAED,MAAM,CAAC,EAAE,EAAE,MAAM,GAAG,OAAO,CAK1B;IAED,IAAI,IAAI,OAAO,CAEd;IAED,0CAA0C;IAC1C,OAAO,SAEN;IAED,6EAA6E;IACvE,IAAI,kBAET;YAEa,OAAO;CAoCtB"}

package/dist/notebook/jobs.js

@@ -0,0 +1,84 @@
+import { randomUUID } from "node:crypto";
+import { errorCode, SemanticPluginError } from "../errors.js";
+/**
+ * Long work as jobs (D11): the action returns a job id at once and the frame polls `jobs.get`.
+ * Jobs of one notebook run one after another; each has its own `AbortController`.
+ */
+export class JobRunner {
+    db;
+    controllers = new Map();
+    chains = new Map();
+    constructor(db) {
+        this.db = db;
+    }
+    start(notebookId, kind, sourceId, work, options = {}) {
+        const id = randomUUID();
+        this.db.insertJob(id, notebookId, kind, sourceId);
+        const controller = new AbortController();
+        this.controllers.set(id, controller);
+        const previous = this.chains.get(notebookId) ?? Promise.resolve();
+        const run = previous.then(() => this.execute(id, work, controller, options));
+        const tail = run.finally(() => {
+            if (this.chains.get(notebookId) === tail)
+                this.chains.delete(notebookId);
+        });
+        this.chains.set(notebookId, tail);
+        return id;
+    }
+    cancel(id) {
+        const controller = this.controllers.get(id);
+        if (!controller)
+            return false;
+        controller.abort();
+        return true;
+    }
+    busy() {
+        return this.controllers.size > 0;
+    }
+    /** Cancels every job (plugin dispose). */
+    dispose() {
+        for (const controller of this.controllers.values())
+            controller.abort();
+    }
+    /** Resolves when every job queued so far has settled (tests and dispose). */
+    async idle() {
+        await Promise.all([...this.chains.values()]);
+    }
+    async execute(id, work, controller, options) {
+        const signal = controller.signal;
+        try {
+            if (signal.aborted)
+                throw new SemanticPluginError("cancelled");
+            this.db.updateJob(id, { state: "running" });
+            let last = 0;
+            const result = await work({
+                signal,
+                progress: (value) => {
+                    const next = Math.max(last, Math.min(1, value));
+                    if (next - last >= 0.01 || next === 1) {
+                        last = next;
+                        this.db.updateJob(id, { progress: next });
+                    }
+                },
+            });
+            if (signal.aborted)
+                throw new SemanticPluginError("cancelled");
+            this.db.updateJob(id, { state: "done", progress: 1, result: result ?? null });
+        }
+        catch (error) {
+            const code = signal.aborted ? "cancelled" : errorCode(error);
+            if (code === "cancelled") {
+                await Promise.resolve(options.onCancel?.()).catch(() => { });
+                this.db.updateJob(id, { state: "cancelled", error: "cancelled" });
+            }
+            else {
+                await Promise.resolve(options.onFail?.(code)).catch(() => { });
+                this.db.updateJob(id, { state: "failed", error: code });
+            }
+        }
+        finally {
+            this.controllers.delete(id);
+        }
+    }
+}
+//# sourceMappingURL=jobs.js.map

package/dist/notebook/jobs.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"jobs.js","sourceRoot":"","sources":["../../src/notebook/jobs.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAC;AACzC,OAAO,EAAE,SAAS,EAAE,mBAAmB,EAAE,MAAM,cAAc,CAAC;AAe9D;;;GAGG;AACH,MAAM,OAAO,SAAS;IACH,EAAE,CAAa;IACf,WAAW,GAAG,IAAI,GAAG,EAA2B,CAAC;IACjD,MAAM,GAAG,IAAI,GAAG,EAAyB,CAAC;IAE3D,YAAY,EAAc;QACxB,IAAI,CAAC,EAAE,GAAG,EAAE,CAAC;IACf,CAAC;IAED,KAAK,CACH,UAAkB,EAClB,IAAY,EACZ,QAAuB,EACvB,IAAa,EACb,OAAO,GAAe,EAAE;QAExB,MAAM,EAAE,GAAG,UAAU,EAAE,CAAC;QACxB,IAAI,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,QAAQ,CAAC,CAAC;QAClD,MAAM,UAAU,GAAG,IAAI,eAAe,EAAE,CAAC;QACzC,IAAI,CAAC,WAAW,CAAC,GAAG,CAAC,EAAE,EAAE,UAAU,CAAC,CAAC;QACrC,MAAM,QAAQ,GAAG,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,UAAU,CAAC,IAAI,OAAO,CAAC,OAAO,EAAE,CAAC;QAClE,MAAM,GAAG,GAAG,QAAQ,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,EAAE,IAAI,EAAE,UAAU,EAAE,OAAO,CAAC,CAAC,CAAC;QAC7E,MAAM,IAAI,GAAG,GAAG,CAAC,OAAO,CAAC,GAAG,EAAE;YAC5B,IAAI,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,UAAU,CAAC,KAAK,IAAI;gBAAE,IAAI,CAAC,MAAM,CAAC,MAAM,CAAC,UAAU,CAAC,CAAC;QAC3E,CAAC,CAAC,CAAC;QACH,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,UAAU,EAAE,IAAI,CAAC,CAAC;QAClC,OAAO,EAAE,CAAC;IACZ,CAAC;IAED,MAAM,CAAC,EAAU;QACf,MAAM,UAAU,GAAG,IAAI,CAAC,WAAW,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAC5C,IAAI,CAAC,UAAU;YAAE,OAAO,KAAK,CAAC;QAC9B,UAAU,CAAC,KAAK,EAAE,CAAC;QACnB,OAAO,IAAI,CAAC;IACd,CAAC;IAED,IAAI;QACF,OAAO,IAAI,CAAC,WAAW,CAAC,IAAI,GAAG,CAAC,CAAC;IACnC,CAAC;IAED,0CAA0C;IAC1C,OAAO;QACL,KAAK,MAAM,UAAU,IAAI,IAAI,CAAC,WAAW,CAAC,MAAM,EAAE;YAAE,UAAU,CAAC,KAAK,EAAE,CAAC;IACzE,CAAC;IAED,6EAA6E;IAC7E,KAAK,CAAC,IAAI;QACR,MAAM,OAAO,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC;IAC/C,CAAC;IAEO,KAAK,CAAC,OAAO,CACnB,EAAU,EACV,IAAa,EACb,UAA2B,EAC3B,OAAmB;QAEnB,MAAM,MAAM,GAAG,UAAU,CAAC,MAAM,CAAC;QACjC,IAAI,CAAC;YACH,IAAI,MAAM,CAAC,OAAO;gBAAE,MAAM,IAAI,mBAAmB,CAAC,WAAW,CAAC,CAAC;YAC/D,IAAI,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,CAAC,CAAC;YAC5C,IAAI,IAAI,GAAG,CAAC,CAAC;YACb,MAAM,MAAM,GAAG,MAAM,IAAI,CAAC;gBACxB,MAAM;gBACN,QAAQ,EAAE,CAAC,KAAK,EAAE,EAAE;oBAClB,MAAM,IAAI,GAAG,IAAI,CAAC,GAAG,CAAC,IAAI,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,KAAK,CAAC,CAAC,CAAC;oBAChD,IAAI,IAAI,GAAG,IAAI,IAAI,IAAI,IAAI,IAAI,KAAK,CAAC,EAAE,CAAC;wBACtC,IAAI,GAAG,IAAI,CAAC;wBACZ,IAAI,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,EAAE,EAAE,QAAQ,EAAE,IAAI,EAAE,CAAC,CAAC;oBAC5C,CAAC;gBACH,CAAC;aACF,CAAC,CAAC;YACH,IAAI,MAAM,CAAC,OAAO;gBAAE,MAAM,IAAI,mBAAmB,CAAC,WAAW,CAAC,CAAC;YAC/D,IAAI,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,EAAE,EAAE,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,CAAC,EAAE,MAAM,EAAE,MAAM,IAAI,IAAI,EAAE,CAAC,CAAC;QAChF,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,GAAG,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC,WAAW,CAAC,CAAC,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC;YAC7D,IAAI,IAAI,KAAK,WAAW,EAAE,CAAC;gBACzB,MAAM,OAAO,CAAC,OAAO,CAAC,OAAO,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC,KAAK,CAAC,GAAG,EAAE,GAAE,CAAC,CAAC,CAAC;gBAC5D,IAAI,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,EAAE,EAAE,KAAK,EAAE,WAAW,EAAE,KAAK,EAAE,WAAW,EAAE,CAAC,CAAC;YACpE,CAAC;iBAAM,CAAC;gBACN,MAAM,OAAO,CAAC,OAAO,CAAC,OAAO,CAAC,MAAM,EAAE,CAAC,IAAI,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,EAAE,GAAE,CAAC,CAAC,CAAC;gBAC9D,IAAI,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,EAAE,EAAE,KAAK,EAAE,QAAQ,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;YAC1D,CAAC;QACH,CAAC;gBAAS,CAAC;YACT,IAAI,CAAC,WAAW,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC;QAC9B,CAAC;IACH,CAAC;CACF"}

package/dist/notebook/ooxml.d.ts

@@ -0,0 +1,6 @@
+import type { Segment } from "./chunk.js";
+/** One segment per slide, in presentation order; notes follow the slide text after a blank line. */
+export declare function pptxSegments(bytes: Buffer): Segment[];
+/** One segment per sheet, in workbook order: "Sheet: <name>" then tab-separated rows. */
+export declare function xlsxSegments(bytes: Buffer): Segment[];
+//# sourceMappingURL=ooxml.d.ts.map

package/dist/notebook/ooxml.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"ooxml.d.ts","sourceRoot":"","sources":["../../src/notebook/ooxml.ts"],"names":[],"mappings":"AAGA,OAAO,KAAK,EAAE,OAAO,EAAE,MAAM,YAAY,CAAC;AA2C1C,oGAAoG;AACpG,wBAAgB,YAAY,CAAC,KAAK,EAAE,MAAM,GAAG,OAAO,EAAE,CAwBrD;AAeD,yFAAyF;AACzF,wBAAgB,YAAY,CAAC,KAAK,EAAE,MAAM,GAAG,OAAO,EAAE,CA2CrD"}